Graft-versus-host disease (GVHD) is a substantial cause of mortality. We hypothesize that conditioning injury creates a pro-inflammatory environment that recruits Teffector cells (Teffs) to the gut, essential for GVHD. Our studies point to the vitamin A metabolite, all-trans-retinoic acid (RA), as a key component of acute GVHD lethality. We have identified critical vitamin A metabolizing enzymes that lead to RA production as markedly up-regulated in the gut and uncovered a fundamental insight that donor T cells sense excessively high RA levels early post-transplant. We show that donor T cells expressing a dominant-negative RAR-alpha (dnRAR?) transgene have markedly impaired GVHD lethality, providing the previously unrecognized cause-and-effect evidence for RA signaling in acute GVHD. Our goal is to determine the best drug approach to block RA production and response. RA signaling occurs via distinct RAR isoforms (?, , ?). Because dnRAR? can sequester its dimeric RXR partners, these data do not prove that targeting RAR? alone was responsible for the striking benefits of GVHD inhibition. This is critical since we now show that RAR? and have distinct functions on in vitro RA induced T cell gut homing receptor expression. Although the RA pathway is highly druggable, no one knows the appropriate RA pathway target. We will solve this dilemma, which will result in the testing of a new class of drugs to restore RAR signaling to homeostatic levels. RA also has anti-inflammatory properties by acting on immune suppressive Tregulatory cells (Tregs). We need to minimize Treg impact. We have shown RA promotes Treg generation, stability and gut homing. An efficient way to expose Tregs to RA in vitro prior to infusion, imprinting Tregs to acquire these desirable properties before transfer. This would favor Tregs over Teffs even in the presence of drugs that inhibit RA production or signaling in vivo. If positive, we have 2 clinical GVHD trials using ex vivo expanded natural and inducible Tregs that could immediately benefit from this strategy. We will test the hypotheses that: Aim 1. High in vivo levels of key vitamin A metabolizing enzymes and RA augment GVHD, whereas in vitro RA exposure augments Treg mediated GVHD suppression regardless of RA signaling in vivo. Hypotheses. We will test the hypothesis that deletion of vitamin A metabolizing enzymes in lineage-specific cell types reduces GVHD, providing insights as to how RA production in situ causes GVHD (1A). We will test the hypothesis that in vitro RA increases Treg generation, stability and gut homing, leading to reduced gut injury and GVHD that can occur in the absence of in vivo RA signaling (1B); Aim 2. RAR isoforms differentially regulate GVHD in vivo. Hypothesis. We will test the hypotheses that RAR isoforms are differentially expressed and have unique functions in donor Teffs, donor Tregs and antigen-presenting cells (2A) and dnRAR? T cells have reduced Teff differentiation, homing and survival (2B). Lastly, we will determine whether our preferred GVHD approach(es) affect graft-versus-leukemia (GVL) responses. When done, we will have developed new therapies to prevent GVHD and maintain GVL.